WO2000015758A2 - Plant resistance to insect pests mediated by viral proteins - Google Patents

Abstract

The invention involves combining a peptide toxin effective against insects, including but not limited to thrips, leaf hoppers, and beetles, with a transport peptide capable of facilitating transfer of the peptide toxin from the gut of an insect to the hemocoel. The combination can be effected by a fusion of genetic material encoding the peptide toxin and the transport peptide, such that expression of the genetic material fusion results in synthesis of a fusion protein combining the functions of both the toxin and the transport protein. Ingestion of the fusion protein by the sucking insect transfers the fusion protein into the insect's gut from which it is transferred into the hemocoel due to the functional activity of the transport peptide where the toxin exerts its toxic effect upon the insect. In a preferred embodiment, the invention is effective in control of such sucking insects as aphids, whiteflies and the like, and other vectors that transmit viruses in a circulative manner.

Description

PLANT RESISTANCE TO INSECT PESTS MEDIATED BY VIRAL PROTEINS

CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority from U.S. Provisional Application No. 60/100,132 filed September 14, 1998.

ACKNOWLEDGMENT OF FEDERAL RESEARCH SUPPORT

This invention was made, at least in part, with funding from the United States Department of Agriculture under contract No. 97-34340-3987. Accordingly, the U.S. government may have certain rights in this invention.

BACKGROUND OF THE INVENTION

Most plant viruses are transmitted by insect vectors. Others are transmitted by mites, fungi or nematodes. Viruses that infect insects, such as baculoviruses, also enter the hemocoel of the insect from the gut through the action of virus coat proteins. Circulatively transmitted plant viruses enter the hemocoel (body cavity) of the insect vector. Viruses that infect insects also enter the hemocoel. In all cases, they have evolved specialized proteins to ensure their survival in the vector. This invention intends to exploit these viral proteins, and the genes encoding them, for presentation and delivery of toxins or other proteins to the vector or related organism. This may apply to non- vectors as well as vectors because, in some cases, viruses often interact with and enter the hemocoel of non- vectors. Furthermore, the viral proteins can be modified to interact with a wider range of organisms than occurs in nature. By facilitating efficient delivery of toxins, this invention can be used to genetically engineer pest-resistant plants, or to construct biopesticides. Plant viruses can be transmitted in a circulative or noncirculative fashion. We define circulative transmission broadly as "any plant virus that must be actively transported across vector membranes and survive inside the vector to be transmitted" (Gray, S.M. [1996] Trends in Microbiology 4:259-264). By this definition, even some fungi can be considered circulatively transmitting vectors. In general, viruses are understood in the art as classifiable within groups which share common features of genetics, structure and the like. Examples of virus groups that are transported across insect vector membranes include: tospoviruses, plant reoviruses, plant rhabdoviruses, tenuiviruses, marafiviruses, luteoviruses, geminiviruses, enamoviruses, tymoviruses, como viruses, and sobemoviruses (ibid: Gergerich, R.C. and Scott, H.A. [1991] Advances in Disease Vector Research. Harris, K.F., ed., pp. 114, Springer- Verlag). The fungally transmitted viruses include bymoviruses and furoviruses (op cit. ; Jianping et al. [1991] Annals of Applied Biology 118:615-622). This invention encompasses all such viruses and their vectors and non- vectors which can acquire the virus.

The term "luteovirus" is used herein to encompass all viruses of the family Luteoviridae, including three major genera: Luteovirus, Polerovirus and Enamovirus. The RPV strain of Barley yellow dwarf virus (BYDV) described in examples herein, is also termed "cereal yellow dwarf virus - RPV" and classified in the genera Polerovirus, as the result of a recent change in nomenclature. However, since the majority of relevant art refers to BYDV, that nomenclature is retained herein.

Noncirculatively transmitted viruses "associate with the cuticular lining of the insect mouthparts or foregut and are released as the insect expels digestive secretions into the plant when it begins to feed. These viruses are not actively transported across vector-cell membranes, nor are they carried internally. The external cuticular lining of insects (and nematodes) extends well into the mouthparts and foregut, but is shed when the animal molts. " (Gray, S.M. [1996] Trends in Microbiology 4:259-264). Although the application herein exemplifies the use of plant virus proteins to deliver toxins to the hemocoel, noncirculatively transmitted viruses may be used similarly to present toxins to the surfaces of the gut or mouthparts. In principle, any virus which can enter the hemocoel of an insect from the gut through the action of viral protein can be employed, as described, to deliver a toxin to the insect. The invention is exemplified herein by luteoviruses and their aphid vectors. They have the best-characterized circulatively-transmitted virus-vector interactions. However, the same principles can be applied for any virus-vector interaction and their exploitation to develop insect-resistant plants is included in this invention.

Luteoviruses can be transmitted only by aphids. The transmission mechanism is persistent and circulative. The virus enters the body cavity (hemocoel) of the aphid where it can remain for the life of the aphid. The aphid acquires virus by feeding on an infected plant. The virus particles (virions) are transported from the aphid hindgut into the hemocoel by a presumed receptor-mediated process across the hindgut epithelial cells. From the hemocoel, the virus is then transported across two more membrane barriers into the accessory salivary gland (ASG) (Gildow, F.E. et al. [1993] Phytopathology 83: 1293-1302; Power, A.G. et al. [1995] in Barlev Yellow Dwarf: 40 Years of Progress. St. Paul: APS Press, pp. 259-289), (Figs. 1 ,2). Subsequently, each time the aphid feeds, it transmits virus by secreting virus-laden saliva into the plant cells.

There is a high degree of vector specificity for different strains of luteoviruses (Power,

A.G. et al. [1995] supra). However, many luteoviruses are transported across the hindgut membrane into the hemocoel, in many nonvector aphid species, (Gildow, F.E. et al. [1993] supra). Hence, vector specificity is thought to be enforced at the ASG barrier. Other viruses, such as potato leafroll virus, are transmitted across the midgut membrane. Nonluteoviruses are digested or excreted without uptake into the hemocoel. Thus, the hindgut epithelial receptors seem to be specific for luteoviruses (Fig. 2). This fact is exploited in the present invention.

The genome of luteoviruses consists of a single, positive sense, 5.7 kb RNA. The proteins needed for virus particles, aphid transmission, and virus movement within the plant all are expressed from a subgenomic RNA (sgRNAl) that is generated during virus infection but is not encapsidated in the virion (Fig. 3) (Miller, W.A. et al. [1997] Plant Disease 81:700- 710). The luteovirus virion contains 180 copies of the coat protein (CP). About 5-10% of the CP subunits contain a long carboxy-terminal extension that protrudes from the virion (Filichkin, S.A. et al. [1994] Virology 205:290-299). This extension arises when ribosomes read through the stop codon of the CP open reading frame (ORF) during translation of sgRNAl (Fig. 3), allowing translation of the downstream ORF5, resulting in a fused CP-readthrough domain (RTD) product (Brown, CM. et al. [1996] J. Virol. 70:5884-5892).

The middle of the RTD (around amino acid 242) contains a labile peptide bond causing the C-terminal half to be cleaved in purified virus preparations (Filichkin et al. [1994] supra). Because purified virions are readily aphid transmitted, the C-terminal portion is unnecessary for aphid transmission. On the other hand, the N-teιτninal half of the RTD is required for aphid transmission (from aphid to plant) (Brault, J. et al. [1995] EMBO J. 14:650-659; van den Heuvel, J.F.J.M. et al. [1997] J. Virol. 71:7258-7265), but not for virion assembly or transport across the hindgut membrane into the hemocoel (Chay, CA. et al. [1996] Virology 219:57-65).

The major CP itself is also required for transmission because intact virions are necessary to protect the viral RNA. In the hemocoel, the N-terminal half of RTD is bound by an abundant protein called symbionin which is produced by endosymbiotic bacteria (van den

Heuvel et al.[1997] supra; Filichkin, S.A. et al. [1997] J. Virol. 71:569-579). This protein strongly resembles bacterial chaperonin groEL which ensures correct folding of many different proteins. However, the binding by symbionin does not resemble that of groEL with its substrate proteins (Hogenhout, S.A. et al. [1998] J. Virol. 72:348-365). The ability of symbionin to bind the luteovirus virion correlates with increased half-life of the virion in the hemolymph (van den Heuvel et al. [1997] supra).

It has been found that a sequence just 3 ' of the CP ORF stop codon (proximal RT element) and a sequence located, surprisingly, 700-750 bases further downstream (distal RT element) are necessary for readthrough of the CP ORF stop codon during translation of sgRNAl (Brown, CM. et al. [1996] supra) (Fig. 3). The distal RT element still facilitates readthrough even after insertion of a reporter gene between it and the proximal element, causing the distal RT element to be located in the 3' untranslated region (UTR) 2 kb downstream from the CP stop codon (Brown, CM. et al. [1996] supra). Current aphid control relies heavily on the use of chemical insecticides. Insecticide application to control transmission of viruses by aphids can have the unintended opposite effect of increasing virus spread by increasing plant-to-plant movement of aphids agitated by sublefhal doses, and by killing aphid predators (Schepers, A. [1989] in Aphids : Their Biology . Natural Enemies and Control. Minks, A.D. et al. [eds.], Elsevier, Amsterdam, Vol C, pp.

123-139). Moreover, the efficacy of chemicals against aphids is limited because of the rapid evolution of insecticide-resistant aphids. Alternative, environmentally benign means of aphid control are required to maintain agricultural productivity.

Use of aphid-resistant crop cultivars has been effective for limiting aphid damage (Auclair, J.L. [1989] in Minks et al., supra, pp. 225-265; Thackray, D.J. et al. [1900] Ann.

Appl. Biol. 116:573-582). Aphid-resistant maize lines (Walter, E.V. et al. [1946] J. Am. Soc.

Agron. 38:974-977) and wheat lines resistant to the Russian wheat aphid (Diuraphis noxia)

(Quisenberry, S.S. et al. [1994] J. Econ. Entomol. 87:1761-1768) have been developed by conventional plant breeding techniques. Transgenic plants that resist insects have been constructed with agents that are active in the gut of insects. The most notable example is the use of the Bacillus thuringiensis toxin (Bt) genes, but no Bt toxin is known that affects aphids.

Transgenic tobacco engineered to express a lectin was shown to confer protection against aphids (Hilder, V.A. et al. [1995] Transgenic Res. 4:18-25), and numerous transgenic plants have been produced that resist aphid-transmitted viruses by the use of virus-derived transgenes (Miller, W.A. et al. [1997] supra; Anon [1995] Genetic Engineering News 15:1; Wilson,

T.M.A. [1993] Proc. Natl. Acad. Sci. 90:3134-3141).

A second approach toward insect pest control has been the use of baculoviruses, which are insect specific viruses. Some of these viruses have been used to deliver a variety of insect- specific toxins that are active in the hemocoel but not in the gut of the insect. For example, recombinant baculoviruses have been developed for control of lepidopteran (moth) pest species

(Bonning, B.C. et al.[1996] - n/2«. Rev. Entomol. 41:191-210). These baculoviruses have been engineered to produce insect hormones such as diuretic hormone (Maeda, S. [1989] Biochem. Biophys. Res. Comm. 165:1177-1183), enzymes such as juvenile hormone esterase (Bonning, B.C. et al. [1997] Proc. Natl. Acad. Sci. USA 94:6007-6012), and insect-specific toxins derived from venomous species such as scorpions and parasitic wasps (McCutchen, B.F. and Hammock, B.D. [1994] in Natural and Derived Pest Management Agents. Hedin, P. et al. [eds.], #551 ed., Washington, D.C: Am. Chem. Soc, pp. 348-367. ACS Symposium Series; Hughes, P.R. et al. [1997] J. Invert. Pathol. 69:112-118; Lu, A. et al. [1996] in Biological control: theory and applications in 1996.7:320: Gershburg, E. et al. [1998] FEBS Lett.

422:132-136; Jarvis, D.L. et al. [1996] in Biological control: theory and applications in 1996.7:228). These insecticidal proteins and peptides cannot be exploited for aphid control at present, because no system exists to deliver them into the hemocoel. Baculoviruses do not infect aphids, and although viruses such as Rhopalosiphum padi virus are known that infect aphids, there is insufficient knowledge of their biology and genetic structure for engineering for aphid control. Furthermore, the release of replicating, transgenic viruses poses greater risk to the ecosystem than expression of nonreplicating viral genes limited to the crop plant, as we disclose here. In the present invention, transgenically expressed, nonreplicating, plant viral structural proteins deliver insecticidal proteins or peptides into the hemolymph of aphids.

The toxin AalT, that is derived from the venom of the North African scorpion

Androctonus australis Hector, has a unique specificity for the nervous system of insects and some other arthropods such as crustaceans (Zlotkin, E. [1986] in Neuropharmacology and pesticide action. Chicester: Horwood; DeDianous S, et al. [1987] Toxicon 25:411-417). AalT has no mammalian toxicity. Its strict selectivity for insects has been documented by toxicity assays, electrophysiological studies and binding assays (Zlotkin [1986] supra). AalT is toxic to all insect species tested. These include over 15 species representing five orders of holo and hemimetabolous insects (Diptera, Coleoptera, Dictyoptera, Orthoptera and Lepidoptera) (Gershburg, E. et al.[1998] supra; Zlotkin [1986] supra; Herrmann, R. et al. [1990] Insect Biochemistry 20:625). The rapid excitatory paralysis induced by AalT results from repetitive firing of the insect' s motor nerves with resulting massive and uncoordinated stimulation of the respective skeletal muscles (Walther, C. et al. [1976] /. Insect Physiol. 22:1187-1194). The insecticidal activity of injected AalT shows that this toxin is among the most potent toxic compounds to insects. AalT has a molecular weight of only 8 kDa and has four disulfide bridges giving a highly organized conformation (Darbon, H. et al. [1982] Int. J. Peptide Protein Res. 20:320-332). AalT does not penetrate the lipophilic cuticle of insects (DeDianous, S. et al. [1988] Pestic. Sci. 23:35-40). To exploit this toxin for insect pest control purposes, various baculoviruses have been engineered to express AalT (Gershburg, E. et al. [1998] supra; Jarvis, D.L. [1996] supra; Maeda, S. et al. [1991] Virology 184:777-780; McCutchen, B.F. et al. [1991] Bio/Technol. 91:848-852; Stewart, L.M.D. et al. [1991] Nature 352:85-88). These viruses infect the tissues of lepidopteran larvae and recombinant AalT produced by the virus is exported from the virus-infected cell into the body cavity, resulting in paralysis of the insect. Expression of AalT is highly effective for control of lepidopteran pest species and significantly reduces the amount of feeding damage caused. Because the nerves (the target site of AalT) of Lepidoptera are unique in being covered by glial cells which act as a barrier to the toxin, AalT is even more toxic to nonlepidopteran species. However, because baculoviruses do not infect aphids, they cannot be used for aphid control without modification. Also, because the site of action of the toxin is the nerves, transgenic plants expressing AalT alone would not be aphid resistant because ingested toxin would pass through the insect.

SUMMARY OF THE INVENTION

The invention involves combining a peptide toxin effective against insects, including but not limited to thrips, leaf hoppers, and beetles, with a transport peptide capable of facilitating transfer of the peptide toxin from the gut of an insect to the hemocoel. The combination can be effected by a fusion of genetic material encoding the peptide toxin and the transport peptide, such that expression of the genetic material fusion results in synthesis of a fusion protein combining the functions of both the toxin and the transport protein. Ingestion of the fusion protein by the sucking insect transfers the fusion protein into the insect's gut from which it is transferred into the hemocoel due to the functional activity of the transport peptide where the toxin exerts its toxic effect upon the insect. In a preferred embodiment, the invention is effective in control of such sucking insects as aphids, whiteflies and the like, and other vectors that transmit viruses in a circulative manner. A variety of transport peptides can be employed, including a coat protein, or transport function domain thereof, of a plant virus for which the sucking insect is a vector in the transmission of the virus from plant to plant. For example, the coat proteins and readthrough domains of certain aphid-transmitted luteoviruses provide transport function in the case of aphids, while the coat-proteins of whitefly-transmitted gemini viruses provide transport function in the case of whiteflies.

Any peptide having a toxic effect when present in the circulatory system of a target insect can, in principle, be incorporated into a toxic fusion protein. Virus proteins that cross the gut barrier of an insect or other pest organism can be exploited for direct delivery of a variety of toxic agents which are active only in the body cavity of that organism. The requirements for these toxic agents include that (1) the agent should be specific for the targeted pest without mammalian toxicity; (2) the agent should be active at low levels; (3) the agent should have a rapid effect on the host. These toxic agents include both toxins that act on the nervous system of insects , and physiological effectors which disrupt regulation of homeostasis in the insect, resulting in feeding inhibition and/or death. A variety of such toxins and physiological effectors have already been exploited specifically for the control of lepidopteran (moth) pests by recombinant baculovirus expression (see Tables 1 and 2 for current lists). The insect-specific neurotoxins have generally been considered to be more effective than physiological effectors, in part because of feed-back regulatory systems in the insect for the latter. There is an ongoing effort within commercial, government and academic laboratories to isolate toxins that specifically target the insect nervous system from a variety of organisms that use venoms to immobilize their prey. The virus coat protein delivery system can be exploited for delivery of all of these agents in an array of pest species.

Table 1. Recombinant baculoviruses expressing neurotoxic peptides.

Origin of toxin Toxin Virus Name Efficacy of virus Lepidopteran Reference

Reduction Reduction host tested in ET₅₀ in feeding

MITE (Tomalski and Miller, 1991;

Pyemotes tritici TxP-l ^a vSp-Tox34 40% Trichoplusia ni Tomalski and Miller, 1992) straw itch mite (Lu et al., 1996) vp6.9tox34 60% T. ni,

Spodoptera frugiperda (Popham et al., 1997)

HzEGTDA 40% Helicopverpa zea

-26tox34

SCORPION

Androctonus AalT AcST-3 10-3! 55-62% T. ni (Cory et al., 1994; Hoover australis AcAalT Heliothis virescens et al., 1995; Maeda et al.,

North African BmAalT Bombyx mori ^c 1991; McCutchen et al.,

(Algerian) scorpion 1991; Stewart et al., 1991) Bonning & Harrison ^d

Ro6.9AaIT 35% Ostrinia nubilalis

Leiurus LqhITI AcLITl .plO 24% Helicoverpa armigera (Gershburg et al., 1998) quinquestriatus LqhIT2 AcLIT2.pol 32% H. armigera (Gershburg et al., 1998) hebraeus BmLqhIT2 B. mori ^c Imai, Aly & Maeda ^d

Israeli yellow LqhIT3 AcLqhIT3 20% H. virescens (Herrmann et al., 1995) scorpion LqhαIT ^b AcLα22 35% H. armigera (Chejanovsky et al., 1995)

SEA ANEMONE

Anemonia sulcata AsII vSAt2p+ 38% 48% T. ni (Prikhod'ko et al., 1998; Prikhod'ko et al., 1996)

Stichadactyla Shi vSShlp+ 36% T. ni (Prikhod'ko et al., 1996) helianthus

SPIDER

Agelenopsis sulcata μ-Aga-IV vMAg4p+ 37% 50% Spodoptera frugiperda (Prikhod'ko et al., 1998;

American funnel T. ni Prikhod'ko et al., 1996) web spider

Tegenaria agrestis TalTX-1 vTalTX-1 18-33 16-39% T. ni (Hughes et al., 1997)

Spodoptera exigua

H. virescens

Diguetia canities DTX9.2 vAcDTX9.2 9-24 30-40% T. ni (Hughes et al., 1997) primitive weaving S. exigua spider H. virescens

TOXIN μ-Aga-IV + vMAg4Sat2 43% 63% T. ni (Prikhod'ko et al., 1998)

COMBINATION AsII S. frugiperda

^a Mechanism of action currently unknown ^b Non-selective neurotoxin. Weak mammalian toxicity. ^c Larvae infected by injection ^d unpublished data

Table 2. Optimization of baculovirus insecticides by expression of physiological effectors.

Agent Virus Target Physiological Efficacy/ Lepidopteran Reference effect Reduction in ST₅₀ host tested

diuretic hormone BmDH5 Malpighian disrupt water 20% B. mori ^a (Maeda, 1989) tubules balance pheromone AcBX-PBAN pheromone unknown 19-16% T. ni (Ma et al., 1998) biosynthesis biosynthesis (5 peptides activating expressed) neuropeptide

(PBAN) juvenile hormone AcJHE-SG unknown disruption of molt 30% T. ni (Bonning et al., esterase (modified) or contraction 66% reduction in H. virescens 1995) paralysis feeding damage

AcJHE-KK juvenile hormone disruption of 50% reduction in H. virescens (Bonning et al., lysosome targeting feeding damage 1997) in pericardial cells chitinase vAcMNPV.chi chitin 22-23% S. frugiperda ^a (Gopalakrishnan et al., 1995) maize URF 13 BV13T mitochondria proteins bind to 40% T. ni ^a (Korth and Levings, BV13.3940 cell membranes 1993) antisense c-myc MycAS cell regulation 15-20% S. frugiperda ^a (Lee et al., 1997) feeding inhibition egt deletion vEGTDEL unclear - deletion degeneration of 20% S. frugiperda (Flipsen et al., of virus egt gene Malpighian 1995a; Flipsen et al., removes virus tubules 1995b; O'Reilly and inactivation of Miller, 1991)

ecdysteroids ^a Larvae injected with virus

All toxins listed in Table 1, (except for TxP-1 and TalTX-1 whose mechanisms of action are currently unknown), disrupt regulation of the sodium channels in the insect nervous system. These neurotoxins act at several different sites on the sodium channel, some being excitatory in action resulting in contractile paralysis, and others inhibitory resulting in slow, flaccid paralysis. Co-injection of insect-selective neurotoxins showed that toxins that act at different sites on the sodium channel can act synergistically (Hermann, R. et al. [1995] Toxicon 33:1099-1102), and hence delivery of multiple toxins to the targeted pest can further improve the pesticidal efficacy of the invention. Furthermore, application of low levels of pyrethroid insecticide (which targets the sodium channels) or a carbamate insecticide (which targets acetylcholinesterase in the nervous system), act synergistically with at least one of the insect specific neurotoxins (McCuthen B.F., [1997] J. Ec. Entomol. 901171-1180). Accordingly, certain chemical insecticides applied at very low doses (LC₁₀-LC₂₀) can also enhance the efficacy of the virus coat protein delivery system when such neurotoxins are used.

The invention allows the person or ordinary skill in the art to construct a fusion protein combining a transport peptide and an insect-toxic peptide for control of a large range of insects that damage a large variety of plants of commercial importance. Such plants include, but are not limited to, wheat, barley, oats, rice, corn (especially maize streak virus in Africa and sweet corn in the US) potato, sugar beet, soybean, tomato, citrus (orange, lemon, lime, grapefruit), Rosaceae (rose), fruit trees (plum, apple, cherry, peach, pear), lettuce, french bean, sugar cane, papaya, squash, cucurbits, banana, cassava, sweet potato, grape, all ornamentals and the like, including other members of the plant families to which the foregoing plants belong.

The family Aleyrodidae (whiteflies), family order Homoptera, including the Aphidae (aphids), family phylloxeridae (the phylloxera pests of grape), the superfamily Fulgoroidae (plant hoppers, the family cicadellidae (leaf hoppers), the Order Thysanoptera (thrips), and the Order Coleoptera, especially the family Chrysomilidae (leaf beetles) and the Genus Diabrotica

(corn rootworm). With regard to the beetles, e.g. corn rootworm, it is known that the adults can transmit circulative viruses, the transport peptides of which can be used for control of those insects. The fusion protein can be delivered to the target insect by any of a variety of ways. However, a preferred method is to deliver the fusion protein during the natural feeding activity of the insect, from the plant itself. The invention therefore includes plant-expressible gene constructs which can be used to transform a plant such that the plant expresses the fusion protein in its sap or tissues where insect feeding occurs. Transgenic plants, capable of expressing a transport peptide-toxin peptide fusion protein are thereby rendered resistant to the damage caused by insects and the diseases transmitted by them. The plant expressible gene constructs can be expressed either constitutively, or in an inducible manner, such as during a desired developmental stage, in a desired tissue, at a desired time or under desired environmental conditions. Constructs can be expressed from stably integrated transgenes or via transient vectors. Modified plant viruses can also be used as a spray on the surface of the crop plant, where ingestion by the target insect introduces the toxic protein to the insect gut.

The invention herein is described by reference to the combination of a luteovirus coat protein and a scorpion toxin, for aphid control. However, it will be understood that other embodiments and variations , according to the principles taught herein and combined with those known in the art, can be made as part of the present invention. The invention therefore includes delivery of any protein to the hemocoel of any arthropod or other organism (nemotode) that takes up and transmits viruses in a circulative or non-circulative fashion.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 is a diagrammatic view of an aphid in saggital section feeding upon a plant, showing in general the route of transmission of a luteovirus.

Fig. 2 is a diagrammatic view of the process of receptor-mediated endocytosis of a luteovirus from the hindgut to the aphid hemocoel.

Fig. 3 is a diagram of the genome organization of BYDV-PAV.

Fig. 4 is a diagram of different CP-RTD-AalT fusions described herein. Fig. 5 shows the effect of wildtype and AalT-expressing virus on aphid mortality.

Fig. 6 illustrates Northern blot hybridization showing accumulation of wildtype and AalT-expressing viral RNAs in oat protoplasts. Fig. 6A shows ethidium bromide stained gel showing positions of ribosomal RNA. Fig. 6B shows positions of RNA probed with 32 P- labeled BYDV-specific antisense RNA.

DETAILED DESCRIPTION OF THE INVENTION

The term "peptide" is used to refer to any poly-amino acid, without limitation as to size or molecular weight. As used herein, "peptide" includes such terms of common usage in the art as "oligo-peptide," "polypeptide" and "protein. "

The term "transport peptide" is herein defined as that peptide segment which is necessary for transport of a circulatively-transmitted virus from the gut to the hemocoel of an insect. A transport peptide can include all, or a portion of, a virus coat protein or other virus protein and can also include all or part of a readthrough domain. That portion of a coat protein or other virus protein which constitutes a transport peptide is termed a component of the coat or other protein. It will be understood in the art that a specific interaction exists between the transport peptide of a virus and the insect host of the virus. A peptide intended to serve as a transport peptide for a given insect species is obtained from a circulatively transmitted virus that is known to infect that insect, as would be understood in the art.

The term "insect-toxic peptide" refers to any peptide which is toxic to an insect when delivered to the appropriate site of action of the insect. The present invention is directed to toxic peptides which exert their effect when delivered to the hemocoel of the insect. Examples of known insect-toxic peptides are given in Table I; however, the number of such peptides that become known is increasing, and any such peptide can be employed in the present invention.

"Readthrough domain" (RTD) is the term used to denote a DNA coding segment, or open reading frame which is situated downstream of a stop codon, in the direction of translation, and whose presence results in synthesis of a fusion protein composed of amino acids encoded upstream of the stop codon and amino acids encoded downstream of the stop codon. The presence of a RTD is indicated by an increased frequency of synthesis of a protein having amino acids encoded by the ORF downstream of the stop codon (readthrough) compared to the frequency of synthesis where the RTD is not present. In the case of an RTD situated downstream of a coat protein gene, the presence of the RTD results in a portion of viral coat proteins having a C-teπr-inal peptide extension. In the case of BYDV, the RTD provides a convenient means for constructing a fusion protein that includes an insect-toxic peptide. The peptide encoded by the RTD of BYDV is not necessary for function as a transport protein, although it may play a role in transport function.

A segment of coding DNA is "expressed" in vivo or in vitro, if the DNA is transcribed or if the transcription product is translated. Expression can result in synthesis of an mR A or of a protein encoded by the coding DNA.

"Associated with/operatively linked" refer to nucleic acid sequences that are related physically or functionally. For example, a promoter or regulatory DNA sequence is said to be "associated with" a DNA sequence that codes for an RNA or a protein if the two sequences are operatively linked, or situated such that the regulator DNA sequence will affect the expression level of the coding or structural DNA sequence.

A "chimeric gene" is a recombinant nucleic acid sequence in which a promoter or regulatory nucleic acid sequence is operatively linked to, or associated with, a nucleic acid sequence that codes for an mRNA or which is expressed as a protein, such that the regulator nucleic acid sequence is able to regulate transcription or expression of the associated nucleic acid sequence. The regulator nucleic acid sequence of the chimeric gene is not normally operatively linked to the associated nucleic acid sequence as found in nature. A chimeric gene having operatively linked coding and expression control segments is also referred to herein as an "expression cassette. " To "control" insects means to inhibit, through a toxic effect, the ability of insect pests to survive, grow, feed, and/or reproduce, or to limit insect-related damage or loss in crop plants, to "control" insects may or may not mean killing the insects, although it preferably means killing the insects.

To "deliver" a toxin means that the toxin comes in contact with an insect, resulting in toxic effect and control of the insect. The toxin can be delivered in many recognized ways, e.g., orally by ingestion by the insect or by contact with the insect via transgenic plant expression, formulated protein compositions(s), sprayable protein composition(s), a bait matrix, or any other art-recognized toxin delivery system.

A "plant" is any plant at any stage of development, particularly a seed plant.

A "plant cell" is a structural and physiological unit of a plant, comprising a protoplast and a cell wall. The plant cell may be in form of an isolated single cell or a cultured cell, or as a part of higher organized unit such as, for example, plant tissue, a plant organ, or a whole plant.

"Plant tissue" as used herein means a group of plant cells organized into a structural and functional unit. Any tissue of a plant in plants or in culture is included. This term includes, but is not limited to, whole plants, plant organs, plant seeds, tissue culture and any groups of plant cells organized into structural and/or functional units. The use of this term in conjunction with, or in the absence of, any specific type of plant tissue as listed above or otherwise embraced by this definition is not intended to be exclusive of any other type of plant tissue.

A "promoter" is an untranslated DNA sequence upstream of the coding region that contains the binding site for RNA polymerase II and initiates transcription of the DNA. The promoter region may also include other elements that act as regulators of gene expression. A "protoplast" is an isolated plant cell without a cell wall or with only part of the cell wall.

"Regulatory elements" refer to sequences involved in controlling the expression of a nucleotide sequence. Regulatory elements comprise a promoter operably linked to the nucleotide sequence of interest and termination signals. They also typically encompass sequences required for proper translation of the nucleotide sequence.

"Transformed/transgenic/recombinant" refer to a host organism such as a bacterium or a plant into which a heterologous nucleic acid molecule has been introduced. The nucleic acid molecule can be stably integrated into the genome of the host or the nucleic acid molecule can be present as an extrachromosomal molecule. Such an extrachromosomal molecule can be auto-replicating. Transformed cells, tissues, or plants are understood to encompass not only the end product of a transformation process, but also transgenic progeny thereof. A "non- transformed," "non-transgenic," or "non-recombinant" host refers to a wild-type organism, e.g. a bacterium or plant, which does not contain the heterologous nucleic acid molecule.

Because luteovirus CP and CP-RTD's are taken up efficiently into the hemocoel, they make ideal vehicles for transporting proteins that are translationally fused to them. Fig. 1 is useful to visualize the circulative pathway of luteovirus transmission (from Chay et al., [1996] Virol. 219:57-65). HG = hindgut; MG = midgut; ASG = accessory salivary gland; PSG - primary salivary gland.

Fig. 2 is a diagram of the proposed mechanism of receptor-mediated endocytosis of luteoviruses from the hindgut into the aphid hemocoel (from Gray, [1996] Trends Microbiol. 4:259-264). The virus binds to specific receptors on the epithelial cells of the hindgut and is taken up by endocytosis. The virus-containing vesicle then fuses with the basal plasmalemma of the epithelial cell, and the virus is released into the hemolymph of the aphid. We exploit these properties to deliver nonviral proteins known to be toxic only when present in the aphid hemocoel. An example of such a toxin is AalT. This gene fusion results in an effective aphicidal transgene product because the protein is toxic at low levels in the hemolymph (Gershburg, E. et al. [1998] supra; Jarvis, D.L. [1996] supra). Equally important, this approach is unlikely to harm nontarget organisms because the toxin is active only in the insect, and the CP-RTD of luteoviruses is specific for aphids. The toxin is also harmless to mammals. Not only will this strategy provide resistance to damage caused directly by the aphid, but it will effectively provide resistance to numerous aphid-transmitted viruses. For example, a variety of diseases caused both by luteo- and nonluteoviruses would be controlled by death of the aphid vector, Myzus persicae . By using a resistance gene that targets aphid species, the invention will have a significant impact both on control of aphids and on control of aphid-transmitted diseases.

Aphids are among the most economically important pest insects of temperate agriculture. They cause yield losses by direct feeding, by the honey dew produced which encourages the growth of harmful sooty molds and, most importantly, by transmission of plant viruses (Harris, K.F. and Maramorosch, K., eds., [1977] Aphids as Virus Vectors. Academic Press, New York; Sylvester, E.S. [1989] in Aphids. Their biology, natural enemies and control. Minks et al. [eds.]), Amsterdam: Elsevier, vol C, pp 65-88). In Britain losses due to aphids themselves average $150 million/year (Tatchell, G.M. [1989], Crop Protection 8:25- 29). In the U.S. , losses due to the recently introduced Russian wheat aphid Diuraphis noxia, alone were $200 million by 1989 (Burton, R.L. [1989] , "The wheat aphid, " 2nd Annual Report of the Agricultural Research Station, USD A, p. 28). Aphid-transmitted viruses cause losses of a similar order. In the U.S. , typical 5 % yield losses in cereals caused by BYDVs would be valued at nearly $500 million (1989 prices) (Hewings, A.D. et al. [1995], in Barley Yellow Dwarf: 40 Years of Progress. D'Arcy, C.J. et al., eds., APS Press, pp. 75-106). Other aphid transmitted viruses cause even greater losses in a wide variety of crops and trees (Miles, P.W. [1989] in Aphids. Their biology, natural enemies and control. Minks, A.K. et al., eds., Amsterdam: Elsevier, vol C, pp. 23-47). These examples show the immense scale of aphid problems.

Although chemical, genetic, and biological control strategies have met with success, all have their limitations and drawbacks. Many tons of fossil fuel are consumed in the manufacture and application of insecticides that are used specifically to control aphids and the viral diseases they transmit. The widespread use of chemical sprays can harm wildlife. Thus, alternative, environmentally acceptable means of aphid control are needed to maintain both farm incomes and sustainable food production.

Biological control methods are limited to specific aphid-parasite interactions and are most effective on perennial crops. The use of genetic aphid resistance reduces the necessity for chemical control, but natural resistance genes are limited. A transgenic approach to aphid resistance allows rapid introduction of new kinds of genes into popular cultivars without the generations of back-crossing needed for introducing genes by conventional breeding. The same constructs can be used for engineering all crops susceptible to aphid attack, because luteoviruses are taken up into the hemocoel of both vector and nonvector aphid species. The invention provides a new weapon in the battery of strategies that are necessary to avoid loss of control through development of aphid resistance to current control measures.

The DNA sequence and translated amino acid sequence of barley yellow dwarf virus coat protein and readthrough domain are given in Table 3 and in SEQ ID NO:l (nucleotide sequence) and SEQ ID NO: 2 (amino acid sequence). The coat protein has the sequence of the first 198 amino acids encoded. Immediately following the codon for amino acid 200 is a TAG (amber) stop codon followed, immediately by GTA, in the same reading frame, encoding a valine. The coat protein-readthrough fusion extends from amino acids 1-669. In Table 3, nucleotide numbers and corresponding encoded amino acid numbers are presented in pairs separated by a slash mark. For example, in the first line of Table 3, the numbers 31/11 designate nucleotide #31, a C directly under numeral 3, and amino acid #11, encoded by the codon beginning at nucleotide #31 (CGC).

The steps of making and evaluating an embodiment of the invention include the following:

1. Construct and express CP-RTD-AalT fusions:

A. In a replicating BYDV-PAV as a transient expression system. B. In the baculovirus expression system for purification of recombinant proteins. 2. Measure aphid toxicity of CP-RTD-AalT fusion proteins. Determine:

A. Aphicidal efficacy of ingested CP-RTD-AalT fusion proteins.

B. Insecticidal efficacy of injected CP-RTD-AalT fusion proteins.

C. Presence of AalT in the aphid hemocoel.

Table 3

DNA sequence 19 56 b.p. AT GAATTCAGTA . .. CAATT: -AGTGA linear

1 / 1 31 / 11

ATG AAT TCA GTA GGT CGT AGA GGA CCT AGA CGC GCA AAT CAA AAT GGC ACA AGA AGG AGG met asn ser val gly arg arg gly pro arg arg ala asn gin asn gly thr arg arg arg

61 / 21 91 / 31

CGC CGT AGA ACA GTT CGG CCA GTG GTT GTG GTC CAA CCC AAT CGA GCA GGA CCC AGA CGA arg arg arg thr val arg pro val val val val gin. pro asn arg ala gly pro arg arg

121 / 41 151 / 51

CGA AAT GGT CGA CGC AAG GGA AGA GGA GGG GCA AAT TTT GTA TTT AGA CCA ACA GGC GGG arg asn gly arg arg lys gly arg gly gly ala asn phe val phe arg pro thr gly giy

181 / 61 211 / ~71

ACT GAG GTA TTC GTA TTC TCA GTT GAC AAC CTT AAA GCC AAC TCC TCC GGG GCA ATC AAA thr glu val phe val phe ser val asp asn leu lys ala asn ser ser gly ala ile lys

241 / 81 ^" 271 / 91

TTC GGC CCC AGT CTA TCG CAA TGC CCA GCG CTT TCA GAC GGA ATA CTC AAG TCC TAC CAT phe gly pro ser leu ser gin cys pro ala leu ser asp gly ila leu lys ser tyr his

301 / 101 331 / 111

CGT TAC AAG ATC ACA AGT ATC CGA GTT GAG TTT AAG TCA CAC GCG TCC GCC AAT ACG GCA arg tyr lys ile thr ser ile arg val glu phe lys ser h s ala ser ala asn thr ala

361 / 121 391 / 131

GGC GCT ATC TTT ATT GAG CTC GAC ACC GCG TGC AAG CAA TCA GCC CTG GGT AGC TAC ATT gly ala ile phe ile glu leu asp thr ala cys lys gin ser ala leu gly ser tyr ile

421 / 141 451 / 151

AAT CC TTC ACC ATC AGC AAG ACC GCC TCC AAG ACC TC CGG TCA GAG GCA ATT AAT GGG asn ser phe thr ile ser lys thr ala ser lys thr phe arg ser glu ala ile asn gly

481 / 161 511 / 171

AAG GAA TTC CAG GAA TCA ACG ATA GAC CAA TTT TGG ATG CTC TAC AAG GCC AAT GGA ACT lys glu phe gin glu ser thr ile asp gin phe trp me leu tyr lys ala asn gly thr

541 / 181 571 / 191

ACC ACT GAC ACG GCA GGA CAA TTT ATC ATT ACG ATG AGT GTC AGT TTG ATG ACG GCC AAA thr thr asp thr ala gly gin phe ile ile thr met ser val ser leu met thr ala lys

601 / 201 631 / 211

TAG GTA GAC TCC TCA ACA CCG CAA CCA AAA CCT GCA CCG CAA CCA ACA CCA ACC CCC CAG

AMB val asp ser ser thr pro glu pro lys pro ala pro glu pro thr pro thr pro gir.

661 / 221 691 / 231

CCA. ACG CCG GCT CCA CAG CCC ACA CCT CAA CCA ACT CCT GCA CCT GTC CCC AAA AGA TTC pro thr pro ala pro gin pro thr pro glu pro thr pro ala pro val pro lys arg phe

721 / 241 751 / 251

TTC GAG TAT ATC GGA ACT CCT ACC GGT ACA ATC TCG ACT ACA GAG AAC ACT GAC AGT ATA phe glu tyr ile gly thr pro thr gly thr ile ser thr arg glu asn thr asp ser ile

781 / 261 811 / 271

TCT GTC AGC AAG CTC GGT GCA CAG TCG ATG CAG TAC ATT CAG AAT GAG AAA TGT GAA ACG ser val ser lys leu gly gly gin ser met gin tyr ile gii-i asn glu lys cys glu thr

841 / 281 871 / 291

AAA GTC ATC GAT TCC TTT TGG AGC ACT AAC AAC AAC GTT TCT GCG CAA GCA GCT TTC GTT lys v l ile asp ser phe trp ser thr asn asn asn val ser ala gl-n ala ala phe val

901 / 301 931 / 311 TAT CCA GTG CCA GAG GGA TCA TAC AGC GTT AAC ATT TCG TGC GAA GGC TTC CAG TCA GTT tyr pro val pro glu gly ser tyr ser val asn ile ser cys glu gly _ϋhe gin ser val 961 / 321 991 / 331

GAC CAC ATC GGT GGC AAC GAG GAC GGC TAT TGG ATT GGT TTA ATT GCC TAC TCC AAT TCG asp his ile gly gly asn glu aso gly tyr tr ile glv leu ile ala ty* s=r asn se- 1021 / 341 ^" 1051 / 351

TCT GGC GAT AAT TGG GGA GTT GGC AAT TAC AAA GGG TGC AGT TTT AAG AAT TTC TTG GCA ser gly asp asn trp gly val gly asn tyr lys gly cvs ser ohe Ivs asn phe i «=u a¹ a

1081 / 361 1111 / 3^*71

ACC AAC ACT TGG AGA CCA GGC CAC AAA GAT CTC AAG TTG ACT GAT TGC CAG TTC ACA GAT thr asn thr trp arg pro gly his lys asn leu lys leu thr aso cys gin Ohe thr asp

1141 / 381 1171 / 391

GGA CAA ATA GTT GAA AGG GAC GCC GTG ATG TCT TTC CAC GTA AA GCA ACA GGC AAG GAT gly gin ile val glu arg asp ala val met ser phe his val glu ala thr gly lvs asp

1201 / 401 1231 / 411

GCC AGC TTC TAC CTC ATG GCT CCC AAA ACA ATG AAA ACT GAC AAA TAC AAC TAT GTT GTC ala ser phe tyr leu met ala pro lys thr met lys thr asp lys tvr asn tyr val val

1261 / 421 1291 / 431 ^{" "}

TCA TAT GGA GGG TAC ACA AAC AAG CGA ATG GAA TTC GGT ACC ATA TCT GTG ACA TGT GAT ser tyr gly gly tyr thr asn lys arg met glu phe gly thr lie ser val thr cys asp

1321 / 441 1351 / 451

GAA TCC GAT GTT GAG GCA GAA CGA ATA ACA AGG CAC GCT CAA ACG CCC ATA CGT TCT AAA glu ser asp val glu ala glu arg ile thr arg his ala glu thr pro ile arg ser lys

1381 / 461 1411 / 471

CAT ATT CTT GTT TCT GAG CGG TAT GCG CAA CCA TTG CCC ACC ATA GTC AAC CAA GGC TTG his ile leu val ser glu arg tyr ala glu pro leu pro thr ile val asn gin gly leu

1441 / 481 1471 / 491

TGT GAT GTG AAA ACT CCC GAG CAA GAA CAA ACA CTG GTG CAT GAA GAT GAC ACA CAA ACT cys asp val lys thr pro glu gin glu gin thr leu val asp glu asp asp arg gin thr

1501 / 501 1531 / 511

GTT TCT ACT GAA TCT GAT ATA GCA CTC CTG GAG TAT GAG GCT GCA ACA GCT GAG ATT CCG val ser thr glu ser asp ile ala leu leu glu tyr glu ala ala thr ala glu ile pro

1561 / 521 1591 / 531

GAT GCT GAA GAG GAC GTT TTG CCC TCC AAG GAA CAG TTG TCT TCA AAA CCA ATG GAT ACG asp ala glu glu asp val leu pro ser lys glu gin leu ser ser lys pro met asp thr

1621 / 541 1651 / 551

TCT GGC AAT ATA ATA CCA AAA CCC AAG GAA CCT GAA GTA CTT GGG ACA TAC CAA GGA CAG ser gly asn ile ile pro lys pro lys glu pro glu val leu gly thr tyr gin gly gin

1681 / 561 1711 / 571

AAC ATT TAT CCT GAA CAC GTA CCT CCA ATG GCG CGG CAG AAA TTG ACA GAA GCC GCG AAT asn ile tyr pro glu asp val pro pro met ala arg gin lys leu arg glu ala ala asn

1741 / 581 1771 / 591

GCG CCT TCC ACG CTA CTC TAT CAA ACA AGA ACC CCA .AAG AAG AGT GGC AAC TTT TTA TCC ala pro ser thr leu leu tyr glu arg arg thr pro lys lys ser gly asn phe leu ser

1801 / 601 1831 / 611

AGA CTT GTA GAA GCG AAT AGG TCC CCT ACT ACT CCC ACT GCC CCA TCC GTG TCA ACT ACT arg leu val glu ala asn arg ser pro thr thr pro thr ala pro ser val ser thr thr

1861 / 621 1891 / 631

TCA AAC ATG ACA AGG CAG CAG CTC CGG GAG TAC ACT AGG ATT AGA AAT TCC AGC GGA ATC ser asn met thr arg glu gin leu arg glu tyr thr arg ie arg asn ser ser gly ile

1921 / 641 1951 / 651

ACA GCA GCA AAG GCG TAC AAG GCG CAA TTC CAG TGA thr ala ala lys ala tyr lys ala gin phe gin OPA Several factors must be considered regarding properties of the CP-RTD and the genes encoding them: (i) the regions of the genes that are required for expression of CP and RTD; (ii) domains in the CP-RTD proteins that facilitate stability in the gut, import into the hemocoel, and stability in the hemolymph; (iii) whether whole virus particles contribute to survival in the aphid digestive system and import into the hemolymph, or whether the fusion proteins themselves are sufficient; and (iv) the effect of different fusions of CP-RTD to AalT on toxicity of AalT. Various approaches are available for expression and testing of the constructs. One is to express the BYDV- AalT fusions from replicating BYDV RNA in infected cells, followed by purification of the protein. The other is to express fusion proteins using the baculovirus expression system. The former approach has the advantage of serving as a rapid, high-level transient expression system in the plant cell which is the native environment in which the constructs are to be expressed as transgenes. The latter allows expression of the gene from DNA in the nucleus out of its viral context which is the case in transgenic plants, and it allows rapid purification of the protein product for direct assessment of its interaction with the aphid. Both approaches can be performed simultaneously.

1. Expression of CP-RTD-AalT fusion proteins A. BYDV-PAV replication-expression system.

The BYDV-PAV transient expression system allows one to identify CP-RTD-AalT fusions with the greatest toxicity to the aphid. This reduces the number of constructs to be tested prior to stable plant transformation. Fig. 3 is a diagram of genome organization of

BYDV-PAV. Bold lines indicate genomic (g) and subgenomic (sg) RNAs. ORFs are numbered as in Chay et al., (1996) Virol. 219:57-65. Coat protein (CP) and readthrough domain (RTD) are translated from sgRNAl . Portion of RTD required for aphid transmission (AT) is upstream of proteolytic cleavage site (scissors). Other functions: polymerase (POL) and systemic movement protein (MP).

In one embodiment, we removed the signal sequence that specifies secretion of AalT (Maeda, S. et al. [1991] Virology 184:777-780; Adachi, T. et al. [1989] J. Biol. Chem. 264:7681-7685), and placed the 200 nt region that codes for the mature AalT protein into ORF 5 (that encodes the RTD) of pPAV6 (Di, R., et al. [1993] Molec. Plant-Microbe Interactions 6:444-453; Mohan, B.R., et al. [1995] Virology 212:186-195). In vitro transcription of pPAV6 with bacteriophage T7 polymerase yields full-length BYDV-PAV genomic RNA that is infectious in oat protoplasts (Mohan, B.R. et al. [1995] Virology 212: 186-195). Hpa I sites were added to the 5' termini of primers flanking the AalT gene termini in plasmid pTZ-AaHIT (McCutchen, B.F. [1991] supra). Following PCR and Hpa I digestion, the fragment was cloned into the unique H.a. I site in pPAVό, placing the AalT coding region in-frame with ORF5, between the proximal and distal RT elements. Fig. 4 shows examples of CP-RTD- AalT fusions to be tested. Scissors indicate approximate site of proteolytic cleavage of the RTD (Filichkin et al., [1994] Virol. 205:290-299; van den Heuvel et al., [1997] J. Virol. 71:7258-7265). Constructs were made with and without the AalT stop codon and with and without the CP stop codon (Fig. 4A, E). A construct corresponding to Fig. 4 A, having the CP stop codon with an AalT insert flanked by segments of the RTD was designated AaIT6, sequence given in SEQ ID NO:4. A second construct, having the CP stop codon but also containing the AalT stop codon, corresponding to Fig. 4E was designated AalTll, sequence given in SEQ ID NO:6. Translated amino acid sequences for AaIT6 and AaIT7 are shown in

SEQ ID NO: 5 and SEQ ID NO:6, respectively. These are tested for expression. The constructs are fully replicated and translated as has been shown with pPAV6. Transcripts containing the much larger 1.8 kb GUS gene inserted in ORF 5 replicated efficiently in oat protoplasts, and the readthrough required for expression of the RTD-GUS fusion was efficient (Brown, CM. et al. [1996] supra).

Where the entire noncleaved portion of the RTD is required for stability of the fusion protein in hemolymph, we place the AalT gene at the distal 3' end of this region. This is less likely to be disruptive to the aphid transmission function of RTD than an internal insertion. The proteolytic cleavage site in the RTD has been located at approximately amino acid 242 (Filichkin, S.A. [1994] supra; van den Heuvel, J.F.J.M. [1997] supra) which corresponds to base 4237 in ORF5 (Figs. 3,4). The cleavage site may is just downstream of the 3 ' end of the distal readthrough element (base 4219) (Brown CM. et al. [1996] supra). It will be understood that inserting the AalT downstream of the proteolytic cleavage site may result in loss of the AalT from the virion before it reaches the hindgut. We insert the AalT just upstream of the cleavage site so that after proteolytic cleavage, it is located very near the C- teπriinus of the RTD (Fig. 4C,D). PCR mutagenesis methods (Landt, O. et al. [1990] Gene 96: 125-128; Kuipers, O.P. et al. [1991] 19:4558) are used to introduce a convenient restriction site for this construct.

The relative efficacy of various constructs can be assessed, to optimize their activity under different conditions. For example, the activity of AalT can be affected by fusion to viral proteins at its N and C termini as in the constructs in Fig. 4 A,C. The constructs that place the AalT ORF at the end of the mature RTD (Fig. 4 C,D,E,F) increase the likelihood that AalT will retain neurotoxicity, because AalT will have only an N-teπrrinal fusion. For all locations in which AalT is inserted, we describe constructs with and without the AalT stop codon. The presence of the AalT stop codon, however, can result in less RTD being expressed, particularly in constructs with the AalT ORF close to the CP stop codon (Fig. 4E). Thus, optimization of efficacy can involve a trade-off between viral and AalT functions. In addition, increasingly large deletions of the RTD and the CP ORFs can be tested. The results allow one to map both the RNA domains required for readthrough of the CP ORF stop codon, and the protein domains required for stability in the aphid and transport of CP-RTD across the hindgut epithelium. The above-described mapping can more precisely define the sequences that contribute to efficient readthrough, and also define other regions suitable for insertion of an insect-toxic peptide coding sequence.

Protoplast inoculation and virus purification. Infectious RNA transcripts and protoplasts were prepared as described previously

(Koev, G. et l., [1999] J Virol. 73:2876-2885; Mohan, B.R. et al., [1995] Virology 212: 186- 195) . Oat protoplasts were transfected with ~ 15 μg of viral RNA transcript by electroporation using the BTX Electro Square Porator T820 (San Diego, CA) with a single, 6 millisecond pulse at 300 V. For each construct, 6 individual samples were transfected. Protoplasts were harvested 48 hr after electroporation by centrifugation at 600xg for 6 min. The supernatant was removed and the pellets resuspended in 400 μl of 10 mM sodium phosphate buffer, pH 7.0. This suspension was sonicated for 2-3 sec. The sonicated protoplasts were centrifuged at 18,000xg (microcentrifuge) at 4°C for 30 min. The supernatant was transferred to a new 1.5 ml tube and kept on ice at a cold room overnight or up to 48 hr. The supernatant was then microcentrifuged again at 18,000xg (4°C) for 30 min. This centrifugation step was repeated as necessary until the supernatant was sufficiently clean. The supernatant was transferred to a 3 ml ultracentrifuge mbe and the volume increased to 3 ml by addition of 10 mM sodium phosphate buffer, pH 7.0. The mbe was then centrifuged at 90,000 rpm at 4°C for 30 min (Beckman TL100 ultracentrifuge, rotor 66iB). The supernatant was discarded and the pellets resuspended in 80 μl of 10 mM sodium phosphate buffer, pH 7.0. This crude virus preparation was used for feeding aphids.

Membrane feeding assay.

The aphid Rhopalosiphum padi was cultured on oat plants (cultivar Clintland64). Containers for aphid feeding were made from 6 x 15 mm petri dishes with black, opaque surfaces. An 18 mm hole was made in the center of the lower petri plate. Each petri dish may hold up to 150 aphids. The partially purified virus extracts (above) were diluted 1:1 or 1:2 in 50% sucrose and fed to virus-free aphids by placing 80-100 μl of the dilution between two Parafilm membranes which were stretched across the hole in the lower petri plate. The plates were placed > 7 cm above a fluorescent light box covered with a yellow filter. After feeding at room temperature overnight (^~16 hr), the numbers of surviving and dead aphids were counted. Natural mortality increased markedly with longer feeding periods.

An initial hurdle was to develop a satisfactory aphid feeding system in which aphids would feed rather than starving due to an aversion or inability to feed. Aphids fed efficiently and survival was high in the feeding apparatus described above. Other feeding systems, or placing aphids too close to the light source resulted in excessive mortality in the absence of virus or toxin. Secondly, the extract from uninfected protoplasts initially resulted in substantial aphid death, most likely resulting from aversion to feeding caused by a component in the crude cell extract. The problem was overcome by partially purifying the virus by the differential centrifugation steps described above.

In two experiments, significantly more aphids died when feeding on extracts of virus harboring AalT in the readthrough domain of the CP (Fig. 5), than died when feeding on negative control solutions. The negative control solutions included (i) sucrose solution, and (ii) extracts of protoplasts inoculated with wildtype virus. Visual inspection revealed that the cause of death of most of the aphids feeding on AaIT6 (Fig. 4 A) or AalTl 1 (Fig. 4E) virus did not resemble feeding aversion, because many aphids died on the membrane. Feeding aversion results in aphids walking around the petri plate, and dying farther from the membrane. Most importantly, the AalT gene is clearly implicated in aphid death. Many more aphids died when feeding on virus encoding AaIT6 and AalTl 1 , than died when feeding on wildtype virus (from plasmid pPAV6) which differs from AaIT6 and AalTl 1 constructs only by the absence of the AalT gene (Fig. 5). There were no differences between the two dilutions of sucrose used in the feeding solutions tested for each transcript. The data in Fig. 5 were subjected to statistical analysis using 1-way ANOVA followed by Tukeys pairwise multiple comparison test for

Buffer, PAV6, AaIT6 and AalTll (n = 4 for each treatment): data for both 1:1 and 1:2 dilutions with 50% sucrose resulting in 25 % (w/v) and 12.5% sucrose, respectively, final concentration. All comparisons are significantly different at p < 0.05, i.e. survival ranks as follows: buffer treatment > PAV > AaIT6 > AalTll.

AalT mRNA has been observed in northern blot hybridizations. Total RNA was isolated 48 hr after inoculation of cells with the same constructs used for the feeding assays. 48 hr after inoculation, total RNA was extracted as in Koev, et al. (1999) supra, from aliquots of protoplasts inoculated with each of the constructs used in Experiment 2 above. RNA was separated by denaturing agarose gel electrophoresis and stained with ethidium bromide to verify loading of total RNA by visualization of ribosomal RNAs (Fig. 6A). The RNA was then blotted to nylon membrane, probed with ³²P-labeled, BYDV-specific antisense RNA as described previously (Koev, et al. 1999). Radioactive bands were detected with a Phosphorimager (Fig. 6B). Mobilities of genomic RNA (gRNA) and the three subgenomic RNAs (sgRNA) produced during virus infection are indicated. sgRNAl is the mRNA for the CP-RTD-AalT fusion protein. Subgenomic RNAs 3 and 4 are small sgRNAs transcribed from the 3' end of the viral genome. As expected, sgRNAl and gRNA from the AaIT6 and AalTll - infected cells are slightly larger (migrate more slowly) than those from PAV6-infected cells, owing to the insertion of the 210 nt AalT gene in the RTD, (Fig. 6B). Subgenomic RNA1 from AalTό and AalTll comigrates with highly abundant ribosomal RNA which partially obscures it in the northern blot hybridization. However, significant levels were still detected (Fig. 6A, lanes AaIT6 and IAIT11). The presence of AalT mRNA (subgenomic RNA1) suggests that the AalT fusion protein was being expressed. The fact that these levels were slightly lower than wildtype (Fig. 6B, lane PAV6) indicates that aphid mortality was not simply proportional to the amount of virus replication, but was correlated with the presence of the AalT gene.

The higher mortality caused by AalTll compared to AaIT6 can be explained by the absence of a carboxy terminal fusion to the AalT protein in AalTll. AaIT6 has the RTD of the CP fused at both its amino- and carboxy termini (Fig. 4 A and SEQ ID NO: 4), whereas AalTll has its own stop codon (Fig. 4E and SEQ ID NO: 6). Thus AalTll has only the aminoterminal fusion to RTD. Fusing additional amino acid sequences to proteins can cause mis-folding or folding such that the domain of interest is not exposed on the surface. These results imply that most of the RTD (the portion downstream of AalT) is not necessary for delivery of AalT to the hemocoel. This is consistent with previous work showing that the RTD, while necessary for virus transmission by the aphid, is not necessary for delivery of the virus to the hemocoel (Chay, et al., [1996] supra; van den Heuvel, et al., [1997] supra).

These experiments allow one to evaluate the effects of a high ratio of CP:RTD compared to removal of the CP stop codon allowing 100% readthrough for translation of the CP-RTD- AalT fusion. It is also possible that the RTD is not needed for all applications. Mutations that completely prevent readthrough still allow virion uptake into the hemolymph (Chay, et al. [1996] supra) but absence of the RTD has been shown to decrease the half-life of luteovirus virions in the hemolymph (van den Heuvel, et al. [1997] supra). In the case where whole virions are needed for transport into hemolymph, the CP stop codon is necessary and AalT is expressed via readthrough, because 100% readthrough prevents virion assembly (Mohan, B.R. [1995] supra). In cases where only a small peptide within the CP is needed for recognition by a hindgut epithelial receptor and subsequent endocytosis into the hemocoel, unnecessary sequences in the CP can be deleted.

Expression of all constructs is monitored by our standard replication assay in protoplasts using northern blot hybridization to detect viral genomic and subgenomic RNAs (Mohan, B.R. [1995] supra; Rasochova, L. etal. [1996] Molecular Plant-Microbe Interactions 9:646-650), and western blots to detect CP and CP-RTD as well as ELISA to measure virion accumulation (Brown, CM. et al. [1996] supra). Aphid toxicity is determined by feeding aphids on extracts of protoplasts, as previously described.

B. Baculovirus expression system

For measuring the dose response of the CP-RTD-AalT fusion proteins, aphids are fed on purified recombinant fusion proteins produced in a baculovirus expression system. The various CP-RTD-AalT fusions described above (Fig. 4) are expressed from DNA coding only for those ORFs (CP, RTD and AalT) rather than a full replicating BYDV genome. It is preferred to use Autographa calif ornica multicapsid nucleopolyhedrovirus (AcMNPV) and standard techniques (O'Reilly, D.R. et al. [1992] Baculovirus Expression Vectors: A Laboratory Manual. W.H. Freeman and Company, New York) that are known in the art. The fusion protein sequences are inserted into the Bgl II cloning site of pAcMPl (Hill-Perkins, M.S. et al. [1990] J. Gen. Virol. 71:971-976). Insect cells (Spodoptera frugiperda; Sf21: (Vaughn, J.L. et al. [1977] In Vitro 13:213-217)) are cotransfected with each recombinant transfer vector and linearized DNA of the virus AcUWl-PH that contains the lacZ gene (Weyer, U. et al. [1990] J. Gen. Virol. 71:1525-1534;Kitts, P. et al. [1993] Biotechniques 14:810-817). The resulting recombinant viruses express each fusion protein under control of the strong baculovirus late promoter /?6.9 (Bonning, B.C. et al. [1994] /. Gen. Virol. 75:1551- 1556). Recombinant viruses (identified by the absence of β-galactosidase activity in the presence of the substrate X-gal), are purified, amplified and used to infect Sf21 cells for production of fusion proteins (O'Reilly et al. [1992] supra).

The method used for purification of the recombinant fusion proteins from insect cell culmre depends on whether virus particles are produced in the cells. Luteovirus coat protein genes have been expressed previously in baculovirus systems resulting in formation of virus particles in the nuclei of infected insect cells (Tian, T. et al. [1995] Virology 213:204-212; Blanc, S. et al. [1993] Virology 197:283-292). If virus particles are produced in AalT fusion constructs, they will be purified by modification of a purification technique that is used routinely for purification of BYDV: the baculovirus-infected cells are lysed by sonication and the BYDV particles purified by differential centrifugation followed by sucrose gradient centrifugation (Rasochova, L. et al. [1996] supra). Because the recombinant virus particles are empty, they settle higher in the gradient than normal virions. If particle assembly does not occur in the baculovirus expression system, one can purify the recombinant proteins using classical protein purification techniques . Specifically , one can purify the recombinant proteins using a polyhistidine tag and a Probond nickel column (Invitrogen) with imidazole elution. The techniques of polyHis fusion tagging and nickel column chromatography are well-known in the art. Recombinant proteins are detected in column fractions by ELISA or western blot, using an anti-AalT antiserum or anti-CP antisera.

2. Bioassay of fusion proteins for paralytic effects

A. Aphicidal efficacy of ingested CP-RTD-AalT fusion proteins

Rhopalosiphumpadi nymphs (30 per sample) are membrane-fed as previously described on one of the following: (i) virus extracts from protoplasts inoculated with AalT-expressing PAV6 transcripts described in IA above; (ii) purified fusion protein from the baculovirus- expression system (IB above); (iii) recombinant AalT alone; (iv) unmodified BYDV-PAV virions; (v) buffer alone. The feeding samples also contain sucrose as described above. This medium is layered between two stretched sheets of Parafilm in the standard aphid feed assay (Rasochova, L. [1996] supra). Aphids are fed overnight and monitored for mortality while feeding on the membrane. Bioassays are repeated with a second agriculturally important aphid species, Myzus persicae. This species is the major vector of several other economically important luteoviruses including potato leaf roll virus, beet western yellows virus, beet mild yellowing virus, and cucurbit aphid-borne yellows virus, as well as viruses in other groups such as soybean mosaic poty virus.

The feeding studies described above have demonstrated that both of the desired properties of the BYDV-AalT fusions are functional: (i) delivery into the hemocoel, and (ii) neurotoxicity. These two properties can be measured separately if desired. As mentioned previously, certain fusion constructs can alter one or both of these activities. Direct injection of the fusion proteins (section 2B) reveals whether AalT is active as a fusion protein, independent of the transport function provided by the virus proteins. Feeding, followed by analysis of proteins in hemolymph (second 2C) can demonstrate transport into the hemocoel as well as providing an indication of stability in the gut and hemolymph. Data from the described studies can be valuable for optimization purposes.

B. Insecticidal efficacy of injected CP-RTD-AalT fusion proteins As a positive control, one can determine the activity of AalT fused to BYDV proteins by injecting fusion proteins into larvae of the blow fly Sarcophaga falculata. This indicator species is particularly sensitive to scorpion toxins and is useful for evaluation of scorpion venom potency (Zlotkin, E. et al. [1971] Biochimie 53: 1073-1078). Comparison of the amounts of fusion protein required to cause contraction of S. falculata larvae with the amounts of unfused, wildtype AalT required for contraction will indicate the relative toxicity of AalT fused to BYDV proteins.

To measure aphid toxicity, one preferably injects apterous (wingless) aphids (which are easier to inject because of the softer cuticle), rather than alate (winged) aphids. Drummond capillaries (20μl) are pulled and broken with fine forceps to produce a point approximately 10 μm in diameter. Each pulled capillary is used for injection of a single aphid to avoid immune reactions to contaminating hemolymph proteins on the capillary. Aphids are immobilized on a microporous filter by suction through a capillary connected to an aspiration system. Injections are conducted using a microinjection apparatus and a dissecting microscope. The volume injected varies according to the applied pressure, which is standardized, and the internal pressure of each aphid, which is variable. Volumes of 10 to 20 nanoliters can be injected without detriment to the aphid (Gildow, F.E. et al. [1993] Phytopathology 83: 1293- 1302).

C. Detection of AalT in aphids

In order to measure the amount of recombinant fusion proteins or control proteins (unmodified BYDV-PAV virions) fed to aphids in section 2A transported into the hemocoel of the aphid, hemolymph samples are collected from 5 aphids in each treatment group by removal of a leg from each aphid and collection of hemolymph in a microcapillary tube (Chay,

CA. [1996] supra). Hemolymph samples from aphids within each group are pooled, and proteins examined by SDS PAGE followed by western blotting. Recombinant proteins are detected in the hemolymph by using antisera to AalT and/or CP. Hemolymph from aphids fed on AalT alone or on buffer alone and examined using the AalT antiserum provide negative controls for test treatments. Hemolymph from aphids fed on BYDV and examined with the CP antiserum provide a positive control for test treatments.

The assays described in section 2 indicate: (i) the toxicity of AalT fused to BYDV structural proteins; (ii) which ingested fusion protein constructs are optimum for toxicity to aphids and which for reducing probing behavior; (iii) which fusion protein constructs have optimum efficiency for transport into the aphid hemocoel. This information indicates the best constructs suitable for use in production of transgenic plants.

Plants can be rendered transgenic by various techniques known in the art. A "transgenic plant" is one which has been genetically modified to contain and express heterologous DNA sequences, either as regulatory RNA molecules or as proteins. A transgenic plant is genetically modified to contain and express at least one heterologous DNA sequence operably linked to and under the regulatory control of transcriptional control sequences which function in plant cells or tissue or in whole plants. As used herein, a transgenic plant also refers to progeny of the initial transgenic plant where those progeny contain and are capable of expressing the heterologous coding sequence under the regulatory control of the plant-expressible transcription control sequences described herein. Seeds containing transgenic embryos are encompassed within this definition.

When plant expression of a heterologous gene or coding sequence of interest is desired, that coding sequence is operably linked in the sense orientation to a suitable promoter and advantageously under the regulatory control of DNA sequences which quantitatively regulate transcription of a downstream sequence in plant cells or tissue or in planta, in the same orientation as the promoter, so that a sense (i.e. , functional for translational expression) mRNA is produced. A transcription termination signal, for example, a polyadenlyation signal functional in a plant cell, is advantageously placed downstream of the CP-RTD coding sequence, and a selectable marker which can be expressed in a plant, can be covalently linked to the inducible expression unit so that after this DNA molecule is introduced into a plant cell or tissue, its presence can be selected and plant cells or tissue not so transformed will be killed or prevented from growing. Where constitutive gene expression is desired, suitable plant- expressible promoters include the 35S or 19S promoters of Cauliflower Mosaic Virus, the Nos, ocs or mass promoters of Agrobacterium tumefaciens Ti plasmids, and others known to the art.

Where tissue specific expression of the plant-expressible insect resistance coding sequence is desired, the skilled artisan will choose from a number of well-known sequences to mediate that form of gene expression. A useful promoter for expression in plant vascular tissue is obtained from sugarcane bacilliform badna virus. The promoter (ScBV3m) is active in both monocots and dicots (Olzsewski, N. [1997] Plant Mol. Biol.). The promoter sequence is given in Table

4 and SEQ ID NO: 3. Environmentally regulated promoters are also well known in the art, and the skilled artisan can choose from well known transcription regulatory sequences to achieve the desired result.

It is understood that nucleic acid sequences other than that of Table 3, will function as coding sequences synonymous with the exemplified CP-RTD coding sequence. Nucleic acid sequences are synonymous if the amino acid sequences encoded by those nucleic acid sequences are the same. The degeneracy of the genetic code is well known to the art; i.e., for many amino acids, there is more than one nucleotide triplet which serves as the codon for the amino acid; for expression in plant cells or tissue it is desired that codon usage reflect that of plant genes and that CpG dinucleotides be kept low in frequency in the coding sequence. It is also well known in the biological arts that certain amino acid substimtions can be made in protein sequences without affecting the function of the protein. Generally, conservative amino acid substimtions or substimtions of similar amino acids are tolerated without affecting protein function. Similar amino acids can be those that are similar in size and/or charge properties, for example, aspartate and glutamate and isoleucine and valine are both pairs of similar amino acids. Similarity between amino acid pairs has been assessed in the art in a number of ways. For example, Dayhoff et al. (1978) in Atlas of Protein Sequence and Structure, Vol. 5, Suppl. 3, pp. 345-352, which is incorporated by reference herein, provides frequency tables for amino acid substimtions which can be employed as a measure of amino acid similarity. Dayhoff et al. 's frequency tables are based on comparisons of amino acid sequences for proteins having the same function from a variety of evolutionarily different sources.

A plant-expressible transcription and translation regulatory sequence can be operably linked to any promoter sequence functional in plants as understood by the skilled artisan; where a regulatory element is to be coupled to a promoter, generally a truncated (or minimal) promoter is used, for example, the truncated 35S promoter of Cauliflower Mosaic Virus, (CaMV). Truncated versions of other constitutive promoters can also be used to provide CAAT and TATA-homologous regions; such promoter sequences can be derived from those of A. tumefaciens T-DNA genes such as Nos, ocs and mas and plant virus genes such as the CaMV 19S gene. It will be understood that the goals of a skilled artisan will determine the choice of particular transcriptional (and translational) regulatory sequences.

A minimal promoter contains the DNA sequence signals necessary for RNA polymerase binding and initiation of transcription. For RNA polymerase II promoters the promoter is identified by a TATA-homologous sequence motif about 20 to 50 bp upstream of the transcription start site and a CAAT-homologous sequence motif about 50 to 120 bp upstream of the transcription start site. By convention, the skilled artisan often numbers the nucleotides upstream of the transcription start with increasingly large numbers extending upstream of (in the 5' direction) from the start site. Generally, transcription directed by a minimal promoter is low and does not respond either positively or negatively to environmental or developmental signals in plant tissue. An exemplary minimal promoter suitable for use in plants is the truncated CaMV 35S promoter, which contains the regions from -90 to +8 of the 35S gene. Where a minimal promoter is used, it is desired that for high levels of gene expression, transcription regulatory sequences which upregulate the levels of gene expression be operably linked thereto. Such quantitative regulatory sequences are exemplified by transcription enhancing regulatory sequences such as enhancers.

Operably linking transcription and translation regulatory sequences upstream of a promoter functional in a plant cell allows the expression of the CP-RTD-toxin fusion coding sequence operably fused just downstream of the promoter, and the skilled artisan understands spacing requirements and ribosome binding site requirements for translational expression of the coding sequence.

Additionally, or alternatively, expression of the regulated construct can be induced, for example, by treating the transgenic plant or tissue with an inducer suitable for regulating expression of the plant-expressible insect resistance coding sequences of the present invention.

The expression of the CP-RTD-toxin fusion coding sequence can also be regulated by tissue specific transcription regulatory sequences . In particular, the sugarcane bacilliforrn badnavirus promoter (ScBV3m) Table 4, is useful for directing expression of a transgene in phloem tissues of both monocots and dicots. Table 4:

C TGCAGGAAGT TGAAGACAAA AGAAGGTCTT AAATCCTGGC TAGCAACACT GAACTATGCC AGAAACCACA TCAAAGATAT GGGCAAGCTT CTTGGCCCAT TATATCCAAA GACCTCAGAG AAAGGTGAGC GAAGGCTCAA TTCAGAAGAT TGGAAGCTGA TCAATAGGAT CAAGACAATG GTGAGAACGC TTCCAAATCT CACTATTCCA CCAGAAGATG CATACATTAT CATTGAAACA GATGCATGTG CAACTGGATG GGGAGCAGTA TGCAAGTGGA AG.AAAAACAA GGCAGACCCA AGAAATACAG AGCAAATCTG TAGGTATGCC AGTGGAAAAT TTGATAAGCC AAAAGGAACC TGTGATGCAG AAATCTATGG GGTTATGAAT GGCTTAGAAA AGATGAGATT GTTCTACTTG GACAAAAGAG AGATCACAGT CAGAACTGAC AGTAGTGCAA TCGAAAGGTT CTACAACAAG AGTGCTGAAC ACAAGCCTTC TGAGATCAGA TGGATCAGGT TCATGGACTA CATCACTGGT GCAGGACCAG AGATAGTCAT TGAACACATA AAAGGGAAGA GCAATGGTTT AGCTGACATC TTGTCCAGGC TCAAAGCCAA ATTAGCTCAG AATGAACCAA CGGAAGAGAT GATCCTGCTT ACACAAGCCA TAAGGGAAGT AATTCCTTAT CCAGATCATC CATACACTGA GCAACTCAGA GAATGGGGAA ACAAAATTCT GGATCC

A transgenic plant can be produced by any means known to the art, including but not limited to Agrobacterium tumefaciens- ediaied DNA transfer, preferably with a disarmed T- DNA vector, electroporation, direct DNA transfer, and particle bombardment and subsequent selection and regeneration (see Davey et al. [1989] Plant Mol. Biol. 13:275; Walden and Schell [1990] Eur. J. Biochem. 192:563; Joersbo and Burnstedt [1991] Physiol. Plant. 81:256;

Potrykus [1991] Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:205; Gasser and Fraley [1989] Sci. 244: 1293; Leemans [1993] Bio/Technology. 11:522; Beck et al. [1993] Bio/Technology. 11: 1524; Koziel et al. [1993] Bio/Technology. 11:194; Vasil et al. [1993] Bio/Technology. 11: 1533). Techniques are well known to the art for the introduction of DNA into monocots as well as dicots, as are the techniques for culturing such plant tissues and regenerating those tissues. Monocots which have been successfully transformed and regenerated include wheat, corn, rye, rice, oat, barley and asparagus. For efficient regeneration of transgenic plants, it is desired that the plant tissue used in the transformation possess a high capacity to produce shoots. For example, Aspen stem sections have good regeneration capacity. (Devillard, C. Ill et al. [1992] C.R. Acad. Sci. Ser. VIE 314:291-298K; Nilsson et al. [1992] Transgenic Research 1:209-220; Tsai et al. [1994] Plant Cell Rep. 14:94-97) Poplars have been successfully transformed (Wilde et al. [1992] Plant Physiol. 98:114-120).

Techniques for introducing and selecting for the presence of heterologous DNA in plant tissue are well known. For example, A. tumefaciens-mediated DNA transfer into plant tissue, followed by selection and growth in vitro and subsequent regeneration of the transformed plant tissue to a plant is well known for a variety of plants.

Other techniques for genetically engineering plant tissue to contain an expression cassette comprising a promoter and associated transcription regulatory sequences fused to the insect resistance coding sequence and optionally containing a transcription termination region are to be integrated into the plant cell genome by electroporation, co-cultivation, microinjection, particle bombardment and other techniques known to the art. The insect resistance plant expression cassette further contains a marker allowing selection of the expression cassette in the plant cell, e.g., genes carrying resistance to an antibiotic such as kanamycin, hygromycin, gentamicin, or bleomycin. The marker allows for selection of successfully transformed plant cells growing in the medium containing certain antibiotics because they will carry the expression cassette with resistance gene to the antibiotic.

It will be understood by those skilled in the art, that the activity of fusion constructs can be optimized to achieve desired levels of transfer into the hemocoel and of toxin dose. In addition, promoter activity in transgenic plants can be varied using techniques and promoters known in the art, in order to achieve levels of expression optimal for the desired level of aphid control. Such modifications, optimizations and constructs are within the scope of the invention.

Claims

WE CLAIM:

1. A fusion protein consisting essentially of a first polypeptide segment, said first segment being a transport peptide of a circulatively transmitted virus and a second polypeptide segment, said second segment being an insect-toxic peptide.

2. A fusion protein according to claim 1, wherein the circulatively transmitted virus is a virus selected from the groups of viruses that include tospovirus, plant reovirus, plant rhabdovirus, tenui virus, marafivirus, luteovirus, geminivirus, enamovirus, tymovirus, como virus, and sobemo virus.

3. A fusion protein according to claim 1, wherein the virus is selected from the group, luteovirus.

4. A fusion protein according to claim 1, wherein the virus is selected from the group, geminivirus.

5. A fusion protein according to claim 3, wherein the virus is barley yellow dwarf virus.

6. A fusion protein according to claim 1 , wherein the transport peptide is a component of the virus coat.

7. A fusion protein according to claim 3 wherein, the transport protein includes a luteovirus coat protein and at least a portion of the readthrough domain.

8. A fusion protein according to claim 4, wherein the transport peptide is a component of the virus coat.

9. A fusion protein according to claim 7, wherein the luteovirus is barley yellow dwarf virus.

10. A fusion protein according to claim 1 , wherein the insect-toxic peptide is selected from the group TxP-1 of Py emotes tritici, AalT of Androctonus australis, LqhlTl, LghIT2, LghITT3 and Lgh╬▒lt of Leiurus quinquestriatus hebraeus, AsII of Anemonia sulcata, Shi of Stichadactyla helianthus, ╬╝-Aga-IV of Agelenopsis sulcata, TalTX-1 of Tegenaria agrestis, and DTX9.2 of Diguetia canities.

11. A fusion protein according to claim 1 , wherein the toxin is toxic to insects of the order Homoptera.

12. A fusion protein according to claim 1, wherein the toxin is toxic to aphids.

13. A fusion protein according to claim 1, wherein the toxin is toxic to whitefly.

14. A fusion protein according to claim 3, wherein the insect-toxic peptide is AalT of Androctonus australis.

15. A fusion protein according to claim 7, wherein the insect-toxic peptide is AalT of Androctonus australis.

16. A fusion protein according to claim 1 having the amino acid sequence of SEQ ID NO:5.

17. A fusion protein according to claim 1 having the amino acid sequence of SEQ ID NO:7.

18. DNA encoding a fusion protein, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide.

19. DNA according to claim 18, wherein the circulatively transmitted virus is a virus selected from the groups of viruses that include tospovirus, plant reovirus, plant rhabdovirus, tenui virus, marafivirus, luteovirus, geminivirus, enamovirus, tymo virus, como virus, and sobemovirus.

20. DNA according to claim 18, wherein the virus is selected from the group, luteovirus.

21. DNA according to claim 20, wherein the virus is selected from the group, geminivirus.

22. DNA according to claim 20, wherein the virus is barley yellow dwarf virus.

23. DNA according to claim 18, wherein the transport protein contains a component of the virus coat.

24. DNA according to claim 20, wherein the transport protein includes a luteovirus coat protein and at least a portion of a readthrough domain downstream of the coat protein.

25. DNA according to claim 24, wherein the nucleotide sequence encoding the transport peptide has an internal stop codon between a portion encoding the coat protein and a portion encoding the readthrough domain.

26. DNA according to claim 24, wherein the nucleotide sequence encoding the transport peptide lacks a stop codon between the portion encoding the coat protein and the portion encoding the readthrough domain.

27. DNA according to claim 24, wherein the nucleotide sequence encoding the transport peptide includes at least one readthrough element.

28. DNA according to claim 18 , wherein the second nucleotide sequence encodes an insect- toxic peptide selected from the group, TxP-1 of Pyemotes tritici, AalT of Androctonus australis, LqhlTl, LghIT2, LghITT3 and Lgh╬▒lt of Leiurus quinquestriatus hebraeus, AsII of Anemonia sulcata, Shi of Stichadactyla helianthus, ╬╝-Aga-IV of Agelenopsis sulcata, TalTX-1 of Tegenaria agrestis, and DTX9.2 of Diguetia canities.

29. DNA according to claim 18, wherein the insect-toxic peptide is toxic to insects of the order Homoptera.

30. DNA according to claim 18, wherein the insect-toxic peptide is toxic to aphids.

31. DNA according to claim 28, wherein the insect-toxic peptide is toxic to whitefly.

32. DNA according to claim 28, wherein the insect-toxic peptide is AalT of Androctonus australis.

33. DNA according to claim 18, having the nucleotide sequence of SEQ ID NO:4.

34. DNA according to claim 18, having the nucleotide sequence of SEQ ID NO:6.

35. A chimeric gene comprising DNA, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide, and a heterologous promoter sequence operatively linked to the DNA encoding the fusion protein.

36. A chimeric gene according to claim 35, wherein said heterologous promoter sequence is the ScBV3m promoter.

37. A recombinant vector comprising the chimeric gene according to claim 35.

38. A transgenic host cell comprising the chimeric gene according to claim 35.

39. A transgenic plant cell comprising the chimeric gene according to claim 35.

40. A transgenic plant or plant tissue comprising the transgenic plant cell of claim 39.

41. A method of producing a fusion protein that is active against insects, comprising

transforming a host cell with a chimeric gene comprising DNA, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide, and a heterologous promoter sequence operatively linked to the DNA encoding the fusion protein, thereby obtaining a transgenic host cell, and

providing the transgenic cell with conditions suitable for expression of the chimeric gene in the transgenic cell, whereby a fusion protein active against insects is produced.

42. A method according to claim 40, wherein said transgenic host cell is a transgenic plant cell.

43. A method of producing an insect-resistant plant comprising introducing a DNA encoding a fusion protein, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide, into said plant, wherein said DNA is expressible in said plant in an effective amount to control said insect.

44. A method according to claim 43, wherein the insect is a member of the order Homoptera.

45. A method according to claim 43, wherein the insert is an aphid.

46. A method according to claim 43, wherein the insect is a whitefly.

47. A method of controlling an insect comprising delivering to the insect an effective amount of a fusion protein consisting essentially of a first polypeptide segment, said first segment being a transport peptide of a circulatively transmitted virus and a second polypeptide segment, said second segment being an insect-toxic peptide.

48. The method of claim 47, wherein theήnsect is a member of the order Homoptera.

49. A method according to claim 47, wherein the insect is an aphid.

50. A method according to claim 47, wherein the insect is a whitefly.

51. A method of controlling an insect comprising feeding the insect an effective amount of a transgenic plant or plant tissue comprising a transgenic plant comprising a chimeric gene comprising DNA encoding a fusion protein, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide, and a heterologous promoter sequence operatively linked to the DNA encoding the fusion protein.

52. The method of claim 51 , wherein the insect is a member of the order Homoptera.

53. A method according to claim 51, wherein the insect is an aphid.

54. A method according to claim 51, wherein the insect is a whitefly.

55. A method of protecting a plant from an insect, comprising expressing in said plant a chimeric gene comprising DNA, said DNA having a first nucleotide sequence encoding a first polypeptide segment, said first polypeptide segment being a transport peptide of a circulatively transmitted virus, and a second nucleotide sequence encoding a second polypeptide segment, said second segment being an insect-toxic peptide, and a heterologous promoter sequence operatively linked to the DNA encoding the fusion protein.

56. The method of claim 55, wherein the insect is a member of the order Homoptera.

57. A method according to claim 55, wherein the insect is an aphid.

58. A method according to claim 55, wherein the insect is a whitefly.